Security and disinfection integrated big data platform regulation and control method based on artificial intelligence

Abstract

The invention discloses a security and disinfection integrated big data platform regulation and control method based on artificial intelligence, and relates to the technical field of safety production and fire-fighting management, and the method comprises the steps: obtaining operation data and historical data, carrying out the standardization processing of the operation data, generating a training data set, training and constructing a deep convolutional neural network model, spatial correlation features are extracted through multilayer convolution operation, key state information is identified in combination with an attention mechanism, and a feature mapping matrix is generated; inputting the feature mapping matrix into a constructed safety evaluation model, calculating a safety risk index, and dividing risk levels according to the safety risk index; and generating an inspection task list according to a risk grade division result, and issuing the inspection task list to the mobile terminal for execution, thereby realizing dynamic regulation and control of the security risk. The technical bottlenecks of a traditional security and disinfection integrated platform in the aspects of data analysis, risk identification, management and control and the like are broken through, intellectualization and precision of security risk management are achieved, and the security management efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of work safety and fire management, and particularly to a control method for an integrated work safety and fire big data platform based on artificial intelligence. Background Art

[0002] Traditional work safety and fire management models have problems such as information silos, scattered data, and low management efficiency, making it difficult to meet the needs of modern management. In recent years, integrated work safety and fire management platforms based on big data technology have gradually emerged. By integrating work safety data and fire monitoring data, the collaborative linkage of safety management and fire management has been realized. However, existing integrated work safety and fire platforms generally have problems such as insufficient data analysis capabilities, untimely risk warnings, and low intelligent levels. Specifically, first, the processing capacity for massive heterogeneous data is limited, making it difficult to fully exploit the deep correlation information contained in the data; second, risk assessment methods rely too much on empirical models and expert knowledge, lacking effective capture of dynamic change characteristics; third, the system response is relatively lagged, unable to achieve real-time perception and intelligent warning of risks.

[0003] To solve the above problems, researchers have begun to try to introduce deep learning technology into the integrated work safety and fire management platform to improve the intelligent level of the system. Currently, some studies have applied convolutional neural networks to safety risk identification and achieved certain results. However, the existing solutions still have the following deficiencies: First, the feature extraction ability is limited, making it difficult to effectively identify complex spatio-temporal correlation features in multi-dimensional data; second, the model generalization ability is insufficient, and the adaptability is poor when facing new risk scenarios; third, there is a lack of effective risk grading mechanisms and dynamic control means, making it difficult to achieve precise control of risks.

[0004] In view of these problems, the present invention proposes a control method for an integrated work safety and fire big data platform based on artificial intelligence. By organically combining a deep convolutional neural network and an attention mechanism, intelligent identification and dynamic control of safety risks are achieved, providing a new technical approach for improving the integrated work safety and fire management level. Summary of the Invention

[0005] In view of the problems existing in the feature extraction ability, model generalization performance, and risk control mechanism of the existing integrated work safety and fire big data platform, the present invention is proposed.

[0006] Therefore, the problem to be solved by the present invention is how to improve the intelligent level of the integrated work safety and fire big data platform through deep learning technology and achieve precise identification and dynamic control of safety risks.

[0007] To solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, an embodiment of the present invention provides a control method for an integrated security and fire protection big data platform based on artificial intelligence, which includes obtaining the operation data and historical data of the integrated security and fire protection big data platform, and performing standardization processing on the operation data to generate a training data set; training and constructing a deep convolutional neural network model based on the training data set, extracting spatial correlation features through multi-layer convolutional operations, and combining an attention mechanism to identify key state information to generate a feature mapping matrix; inputting the feature mapping matrix into the constructed security assessment model, calculating a security risk index, and dividing risk levels according to the security risk index; generating an inspection task list according to the risk level division result and sending it to a mobile terminal for execution to achieve dynamic control of security risks.

[0009] As a preferred solution of the control method for the integrated security and fire protection big data platform based on artificial intelligence of the present invention, it includes: generating an inspection task list according to the risk level division result and sending it to a mobile terminal for execution to achieve dynamic control of security risks, including: generating an inspection task list based on the risk level division result, and allocating inspection resources by using a task priority sorting algorithm, where the inspection task list includes a first-level inspection task list, a second-level inspection task list, and a third-level inspection task list; the inspection task list includes an inspection path, the position coordinates of monitoring points, and inspection frequency requirements; if the security risk level is a fourth-level risk, a third-level inspection task list is generated; if the security risk level is a third-level risk, a second-level inspection task list is generated; if the security risk level is a second-level risk, a first-level inspection task list is generated; pushing the inspection task list to the mobile terminal device of the inspector to carry out inspection work, and collecting image data and environmental parameters of the inspection points through the mobile terminal device, where the mobile terminal device plans the optimal inspection route based on the built-in electronic map navigation module; analyzing the image data and the environmental parameters, and transmitting the analysis result to the integrated security and fire protection big data platform to output inspection feedback data; updating the parameters of the risk assessment model through the inspection feedback data and dynamically correcting the risk level division result.

[0010] As a preferred solution of the artificial intelligence-based integrated security and fire control big data platform regulation method of the present invention, the method includes: inputting the feature mapping matrix into the constructed security assessment model, calculating the security risk index, and dividing the risk levels according to the security risk index, including: based on historical data, constructing a non-linear mapping space using a radial basis kernel function, and selecting kernel function parameters and penalty factors through a cross-validation method to construct a security assessment model; calculating weight coefficients for the spatial features and temporal features in the feature mapping matrix respectively, and constructing a composite feature vector in a weighted combination manner; inputting the composite feature vector into the security assessment model, calculating the distance from the sample point to the classification hyperplane to obtain the security risk index; setting a grading threshold for the security risk index according to the statistical distribution of historical data, calculating the change trend of the security risk index at multiple consecutive time points using a sliding time window, establishing a risk warning level, and setting a warning strategy.

[0011] As a preferred solution of the artificial intelligence-based integrated security and fire control big data platform regulation method of the present invention, the specific formula of the security assessment model is as follows:

[0012]

[0013] Among them, R(t) is the security risk index at time t, α i is the support vector coefficient of the i-th sample, x i is the support vector sample in historical data, K(x i , H(t)) is the improved radial basis kernel function, H(t) is the feature mapping matrix, b is the bias term, λ is the regularization coefficient, w j is the weight coefficient of the j-th feature dimension, F j (t) is the temporal feature integral term of the j-th feature dimension, is the time window smoothing function, n is the number of samples, and m is the number of extracted feature dimensions.

[0014] As a preferred solution of the artificial intelligence-based integrated security and fire control big data platform regulation method of the present invention, the method further includes: when the security risk index R(t) < the first threshold and |R(t k ) - R(t k-1 )| < the preset threshold within a continuous time window, it is determined as a first-level risk; if the first threshold ≤ the security risk index R(t) < the second threshold and the risk index R(t k ) at the current moment k is greater than the risk index R(t k-1), it is determined as a level 2 risk and triggers a level 1 warning; if the second threshold ≤ safety risk index R(t) < the third threshold and the safety risk index R(t) remains in this range within the continuous time window, it is determined as a level 3 risk and triggers a level 2 warning, adjusts the inspection route, dispatches additional professional inspectors, and formulates an emergency plan; when the safety risk index R(t) ≥ the third threshold or within the continuous time window|R(t k )-R(t k-1 )|>When the preset threshold is reached, it is judged as a level 4 risk, triggering a level 3 warning, activating the emergency response mechanism, and implementing special control measures.

[0015] As a preferred solution of the artificial intelligence-based integrated security and firefighting big data platform control method described in the present invention, wherein: the method for generating the feature mapping matrix is ​​to establish a deep convolutional neural network model according to a deep learning framework, and divide the training data set into a training set and a validation set according to an expected ratio; at the same time, a small batch stochastic gradient descent method is used to optimize the training of the deep convolutional neural network model, wherein the deep convolutional neural network model includes an input layer, a multi-layer convolution layer, a pooling layer, a fully connected layer and an output layer; spatial correlation features are extracted through multi-layer convolution operations, a self-attention mechanism is introduced to calculate the weight coefficients between different channels, and the channel attention weight and the spatial attention weight are multiplied to obtain a fused attention weight; the fused attention weight and the original feature map are weighted and summed, and the weighted features are mapped to dimension reduction through a fully connected layer to generate a feature mapping matrix, wherein the dimension of the feature mapping matrix is ​​n×m, wherein n represents the number of samples and m represents the extracted feature dimension.

[0016] As a preferred solution of the artificial intelligence-based integrated safety and firefighting big data platform control method described in the present invention, the training data set is generated by establishing a multi-source data acquisition channel to collect the operating data of the integrated safety and firefighting big data platform through Internet of Things sensors, wherein the operating data includes monitoring data, firefighting facility distribution data and safety inspection data; the monitoring data is cleaned to remove outliers and duplicate values, and a linear interpolation method is used to supplement missing values; at the same time, the firefighting facility distribution data is coordinate unified and scale standardized, and the safety inspection data is time series aligned and format unified; the maximum and minimum value normalization method is used to map all processed data to the [0,1] interval to generate a training data set.

[0017] Second aspect, an embodiment of the present invention provides an integrated safety and fire control big data platform regulation system based on artificial intelligence, which includes: an acquisition module, configured to acquire the operation data and historical data of the integrated safety and fire control big data platform, and perform standardization processing on the operation data to generate a training data set; a feature extraction module, based on the training data set, training and constructing a deep convolutional neural network model, extracting spatial correlation features through multi-layer convolutional operations, and combining an attention mechanism to identify key state information to generate a feature mapping matrix; a safety evaluation and risk calculation module, configured to input the feature mapping matrix into a constructed safety evaluation model, calculate a safety risk index, and divide risk levels according to the safety risk index; a patrol task generation and distribution module, configured to generate a patrol task list according to the risk level division result and distribute it to a mobile terminal for execution to achieve dynamic regulation of safety risks.

[0018] Third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program instructions are executed by the processor, the steps of the integrated safety and fire control big data platform regulation method based on artificial intelligence as described in the first aspect of the present invention are implemented.

[0019] Fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program instructions are executed by the processor, the steps of the integrated safety and fire control big data platform regulation method based on artificial intelligence as described in the first aspect of the present invention are implemented.

[0020] The beneficial effects of the present invention are as follows: By performing standardization processing on operation data and historical data, the problem of heterogeneous data unification is solved; through the combination of a deep convolutional neural network model and an attention mechanism, deep feature extraction of multi-dimensional data and key information recognition are realized, improving the accuracy of risk recognition; through the safety evaluation model of the feature mapping matrix, quantitative evaluation and hierarchical management of risks are realized, providing a scientific basis for risk control; through intelligent generation and real-time distribution of patrol tasks, closed-loop regulation of risk management is realized; the overall solution breaks through the technical bottlenecks of traditional integrated safety and fire control platforms in aspects such as data analysis, risk recognition, and control, realizes the intelligence and precision of safety risk management, and significantly improves safety management efficiency. Description of the Drawings

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts. Among them:

[0022] Figure 1Flowchart of the regulation method for the integrated safety and fire protection big data platform based on artificial intelligence in Embodiment 1. Detailed implementation manners

[0023] To make the above objects, features and advantages of the present invention more obvious and understandable, the following detailed description of the specific implementation manners of the present invention will be given in conjunction with the accompanying drawings of the specification.

[0024] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0025] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure or characteristic that can be included in at least one implementation manner of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that is mutually exclusive with other embodiments.

[0026] Embodiment 1

[0027] Referring to Figure 1 , which is the first embodiment of the present invention. This embodiment provides a regulation method for an integrated safety and fire protection big data platform based on artificial intelligence, including:

[0028] S1: Obtain the operation data and historical data of the integrated safety and fire protection big data platform, and perform standardization processing on the operation data to generate a training data set.

[0029] Specifically, the method for generating the training data set is to establish a multi-source data acquisition channel, and collect the operation data of the integrated safety and fire protection big data platform through Internet of Things sensors, where the operation data includes monitoring data, fire protection facility distribution data, and safety inspection data.

[0030] It should be noted that the monitoring data includes temperature data collected by temperature sensors, smoke concentration data collected by smoke sensors, harmful gas concentration data collected by gas sensors, and pressure data collected by pressure sensors; synchronously collect fire protection facility distribution data, and the fire protection facility distribution data includes fire hydrant location data, fire extinguisher distribution data, emergency lighting equipment distribution data, and fire passage distribution data; collect safety inspection data uploaded by inspection personnel, and the safety inspection data includes inspection route data, inspection point data, inspection item data, and inspection result data.

[0031] Furthermore, clean the monitoring data by removing outliers and duplicate values, and use the linear interpolation method to fill in the missing values. At the same time, perform coordinate unification and scale standardization on the fire facility distribution data, and perform time series alignment and format unification on the safety inspection data.

[0032] Moreover, use the min-max normalization method to map all the processed data to the interval [0, 1] to generate a training dataset.

[0033] S2: Based on the training dataset, train and construct a deep convolutional neural network model, extract spatial correlation features through multi-layer convolutional operations, and combine the attention mechanism to identify key state information to generate a feature map matrix.

[0034] Specifically, the method for generating the feature map matrix is as follows: establish a deep convolutional neural network model according to the deep learning framework, and divide the training dataset into a training set and a validation set according to the expected ratio. At the same time, use the mini-batch stochastic gradient descent method to optimize and train the deep convolutional neural network model, where the deep convolutional neural network model includes an input layer, multiple convolutional layers, pooling layers, fully connected layers, and an output layer.

[0035] It should be noted that the input layer receives the training dataset and reconstructs the data into a three-dimensional tensor form. The first convolutional layer uses a 3×3 convolutional kernel to extract features from the input data and extract low-level spatial features. The second convolutional layer uses a 5×5 convolutional kernel for feature mapping and extracts middle-level semantic features. The third convolutional layer uses a 7×7 convolutional kernel for feature combination and extracts high-level abstract features. A max pooling layer is set between each convolutional layer to compress the feature dimension through a 2×2 sliding window and reduce the computational complexity.

[0036] Furthermore, extract spatial correlation features through multi-layer convolutional operations, introduce the self-attention mechanism to calculate the weight coefficients between different channels, and multiply the channel attention weight and the spatial attention weight to obtain the fused attention weight. Sum the fused attention weight and the original feature map with weights, and map the weighted features through a fully connected layer for dimensionality reduction to generate a feature map matrix, where the dimension of the feature map matrix is n×m, where n represents the number of samples and m represents the extracted feature dimension.

[0037] S3: Input the feature map matrix into the constructed safety assessment model, calculate the safety risk index, and divide the risk levels according to the safety risk index.

[0038] Specifically, based on historical data, use the radial basis kernel function to construct a non-linear mapping space, and select the kernel function parameters and penalty factors through the cross-validation method to construct a safety assessment model.

[0039] Furthermore, the specific formula of the security assessment model is as follows:

[0040]

[0041] Among them, R(t) is the security risk index at time t, α i is the support vector coefficient of the i-th sample, x i is the support vector sample in the historical data, K(x i , H(t)) is the improved radial basis kernel function, H(t) is the feature mapping matrix, b is the bias term, λ is the regularization coefficient, w j is the weight coefficient of the j-th feature dimension,

[0042] F j (t) is the time series feature integral term of the j-th feature dimension, is the time window smoothing function, n is the number of samples, and m is the number of extracted feature dimensions.

[0043] Furthermore, the specific formula of the improved radial basis kernel function is as follows:

[0044]

[0045] Among them, β is the feature correlation adjustment coefficient, and θ is the included angle of the feature vectors.

[0046] It should be noted that the specific formulas of the time window smoothing function and the time series feature integral term are as follows:

[0047]

[0048] Among them, W is the time window value, γ is the time decay coefficient, τ is the integration time range, R(t k ) is the risk index at the current moment k, and R(t k-1 ) is the risk index at the previous moment k - 1.

[0049] Specifically, the weight coefficients of the spatial features and the time series features in the feature mapping matrix are calculated respectively, and a composite feature vector is constructed by using a weighted combination method; the composite feature vector is input into the security assessment model, the distance from the sample point to the classification hyperplane is calculated to obtain the security risk index; according to the statistical distribution of the historical data, the classification threshold of the security risk index is set, and the change trend of the security risk index at multiple consecutive time points is calculated by using a sliding time window to establish a risk warning level and set a warning strategy.

[0050] Furthermore, when the security risk index R(t) < the first threshold and within the continuous time window |R(t k ) - R(t k-1) When it is less than the preset threshold, it is determined as a first-level risk; if the first threshold ≤ safety risk index R(t) < the second threshold and the risk index R(t k ) at the current moment k is greater than the risk index R(t k-1 ) at the previous moment k - 1, it is determined as a second-level risk and a first-level early warning is triggered; if the second threshold ≤ safety risk index R(t) < the third threshold and the safety risk index R(t) remains within this interval within a continuous time window, it is determined as a third-level risk, a second-level early warning is triggered, the inspection route is adjusted, professional inspectors are dispatched, and an emergency plan is formulated; when the safety risk index R(t) ≥ the third threshold or |R(t k ) - R(t k-1 )| is greater than the preset threshold within a continuous time window, it is determined as a fourth-level risk, a third-level early warning is triggered, the emergency response mechanism is activated, and special control measures are implemented.

[0051] It should be noted that the first threshold is determined based on the normal fluctuation range of the safe operation state in historical data; the second threshold is determined based on the distribution value of slight abnormal states in historical data statistical analysis; the third threshold is determined based on the risk accumulation critical value in the early stage of historical major safety accidents; the preset threshold is determined based on the risk accumulation effect and risk diffusion trend within a continuous time window; the measures for the first-level early warning include increasing the monitoring frequency of key areas and adjusting the daily inspection plan; the measures for the second-level early warning include adjusting the inspection route, dispatching professional inspectors, and formulating an emergency plan; the measures for the third-level early warning include activating the emergency response mechanism and implementing special control measures, including evacuating personnel, cutting off the hazard source, and activating fire-fighting facilities.

[0052] S4: Generate an inspection task list according to the risk level classification result, and send it to the mobile terminal for execution to achieve dynamic control of safety risks.

[0053] Specifically, based on the risk level classification result, generate an inspection task list, and use the task priority sorting algorithm to allocate inspection resources; push the inspection task list to the mobile terminal device of the inspector to carry out the inspection work, and the mobile terminal device plans the optimal inspection route based on the built-in electronic map navigation module and collects the image data and environmental parameters of the inspection points.

[0054] It should be noted that the inspection task list includes the first-level inspection task list, the second-level inspection task list, and the third-level inspection task list; the inspection task list includes the inspection path, the location coordinates of the monitoring points, and the requirements for the inspection frequency. The first-level inspection task list corresponds to the second-level risk situation, mainly including increasing the daily inspection frequency, focusing on abnormal areas on the basis of the original inspection path, appropriately densifying the layout of the monitoring point locations, and the inspection frequency requirement is increased from once a day to twice a day; the second-level inspection task list corresponds to the third-level risk situation, and it is necessary to adjust and optimize the inspection path, add temporary monitoring points in key areas, and the location coordinates of the monitoring points should cover all risk hidden danger areas, and the inspection frequency requirement is increased to four times a day, and professional inspectors are required to participate in the inspection; the third-level inspection task list corresponds to the highest-level fourth-level risk situation, and it is necessary to re-plan the emergency inspection path, deploy dense monitoring points in high-risk areas, and the location coordinates of the monitoring points need to achieve full coverage and densification in key areas, and the inspection frequency requirement is increased to once every two hours, and the inspection task is performed by professional emergency personnel.

[0055] Furthermore, if the safety risk level is a fourth-level risk, a third-level inspection task list is generated; if the safety risk level is a third-level risk, a second-level inspection task list is generated; if the safety risk level is a second-level risk, a first-level inspection task list is generated.

[0056] Moreover, the image data and environmental parameters are analyzed, and the analysis results are transmitted to the integrated safety and fire protection big data platform to output inspection feedback data; the parameters of the risk assessment model are updated through the inspection feedback data, and the risk level classification results are dynamically corrected.

[0057] Furthermore, this embodiment also provides an integrated safety and fire protection big data platform control system based on artificial intelligence, including: an acquisition module, which is used to acquire the operation data and historical data of the integrated safety and fire protection big data platform, and perform standardization processing on the operation data to generate a training data set; a feature extraction module, based on the training data set, trains and constructs a deep convolutional neural network model, extracts spatial correlation features through multi-layer convolutional operations, and combines the attention mechanism to identify key state information to generate a feature mapping matrix; a safety assessment and risk calculation module, which is used to input the feature mapping matrix into the constructed safety assessment model, calculate the safety risk index, and divide the risk level according to the safety risk index; an inspection task generation and distribution module, which is used to generate an inspection task list according to the risk level classification result and distribute it to the mobile terminal for execution to achieve dynamic control of safety risks.

[0058] This embodiment also provides a computer device, which is applicable to the situation of the regulation method of the integrated security and fire protection big data platform based on artificial intelligence, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the regulation method of the integrated security and fire protection big data platform based on artificial intelligence as proposed in the above embodiment.

[0059] The computer device can be a terminal. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner. The wireless manner can be achieved through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covered on the display screen, or a button, a trackball, or a touchpad set on the outer shell of the computer device, or an external keyboard, a touchpad, or a mouse, etc.

[0060] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, the following steps are implemented: obtaining the operation data and historical data of the integrated security and fire protection big data platform, and performing standardization processing on the operation data to generate a training data set; based on the training data set, training and constructing a deep convolutional neural network model, extracting spatial correlation features through multi-layer convolutional operations, and combining an attention mechanism to identify key state information to generate a feature mapping matrix; inputting the feature mapping matrix into the constructed security assessment model, calculating a security risk index, and dividing risk levels according to the security risk index; generating an inspection task list according to the risk level division result and sending it to a mobile terminal for execution to achieve dynamic regulation of security risks.

[0061] In summary, the present invention solves the problem of unifying heterogeneous data through the standardization processing of operation data and historical data; through the deep convolutional neural network model combined with the attention mechanism, it realizes the extraction of deep features of multi-dimensional data and the identification of key information, improving the accuracy of risk identification; through the security assessment model of the feature mapping matrix, it realizes the quantitative assessment and hierarchical management of risks, providing a scientific basis for risk control; through the intelligent generation and real-time sending of inspection tasks, it realizes the closed-loop regulation of risk management; the overall solution breaks through the technical bottlenecks in data analysis, risk identification, and control of traditional integrated security and fire protection platforms, realizes the intelligence and precision of security risk management, and significantly improves the efficiency of security management.

[0062] Example 2

[0063] Referring to Table 1, this is the second embodiment of the present invention. This embodiment provides a control method for an integrated safety and fire protection big data platform based on artificial intelligence. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0064] Specifically, deploy an integrated safety and fire protection big data platform system in a large commercial complex with a floor area of 50,000 square meters; install 1,200 Internet of Things sensor nodes in the building, including 500 temperature sensors (model: TS-2000, measurement range: -20°C to 120°C, accuracy: ±0.1°C), 300 smoke sensors (model: SM-500, detection range: 0 to 500 ppm, response time < 10 s), 200 harmful gas sensors (model: GD-300, can detect CO, NO2, etc., detection range: 0 to 100 ppm), and 200 pressure sensors (model: PS-100, range: 0 to 10 MPa). At the same time, accurately locate 297 fire protection facility points in the building, including 120 fire hydrants, 97 fire extinguisher boxes, 50 emergency lighting devices, and 30 fire corridors.

[0065] Furthermore, collect operation data for 90 days, with a data sampling frequency of once every 5 minutes. Through data cleaning, approximately 3.2% of outliers and 1.8% of duplicate values are removed in total, and approximately 2.5% of missing values are supplemented using the linear interpolation method. For the fire protection facility distribution data, the WGS84 coordinate system is unified and standardized at a scale of 1:100. For the safety inspection data, all timestamps are unified to the UTC+8 time zone, and the data format is normalized.

[0066] Even further, construct a deep convolutional neural network model, and divide the training set and validation set in a ratio of 80%:20%; the training of the deep convolutional neural network model uses the mini-batch stochastic gradient descent method with a batch size of 64, the initial value of the learning rate is set to 0.001, and the cosine annealing strategy is used for dynamic adjustment. After reaching an accuracy of 95.8% on the validation set, start the actual application test.

[0067] Specifically, as shown in Table 1, from the perspective of the mean risk index, different functional areas show a reasonable differential distribution; the mean risk index of the equipment machine room area is the highest at 0.578, which is consistent with its characteristics of dense equipment and large power consumption; while the risk index of the commercial office area is the lowest at 0.312, reflecting the relatively stable safety state of this area. This distribution characteristic verifies that the risk assessment model can accurately capture the inherent risk characteristics of different regions.

[0068] Table 1. Test data table

[0069]

[0070] Furthermore, in terms of the early warning accuracy rate, the overall performance of the system is excellent, and the accuracy rate in all regions reaches more than 95%, with the equipment machine room area reaching the highest of 98.1%. It is particularly worth noting that the false negative rate of the system is generally controlled at about 1%, and the false positive rate also remains at a low level. The highest false positive rate in the underground parking lot area is only 1.5%, which greatly reduces unnecessary waste of human resources. The average response time is controlled within 2.5 seconds, and in the equipment machine room area, it only takes 1.9 seconds to complete risk assessment and early warning release. The accuracy of risk level determination all maintains above 96%, fully demonstrating the superiority of the deep convolutional neural network model in feature extraction and risk level classification.

[0071] Even further, remarkable results have been achieved in terms of the inspection task completion rate, and the completion rate in all regions reaches more than 98.5%. This benefits from the intelligent task allocation mechanism based on risk levels and the precise navigation function of the mobile terminal. Although the risk index in the dining area is relatively high at 0.521, through the dynamic regulation of the system, its early warning accuracy rate reaches 97.2% and the inspection task completion rate reaches 99.3%, fully demonstrating the precise control ability of the system in high-risk areas.

[0072] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A control method for an integrated security and firefighting big data platform based on artificial intelligence, characterized by: include, Obtain the operating data and historical data of the integrated security and firefighting big data platform, and standardize the operating data to generate a training data set; Based on the training data set, a deep convolutional neural network model is trained and constructed, spatial correlation features are extracted through multi-layer convolution operations, and key state information is identified in combination with an attention mechanism to generate a feature mapping matrix; Inputting the feature mapping matrix into the constructed security assessment model, calculating the security risk index, and dividing the risk level according to the security risk index; An inspection task list is generated based on the risk level classification results and sent to the mobile terminal for execution, realizing dynamic regulation of security risks.

2. The artificial intelligence-based security and fire protection integrated big data platform control method according to claim 1, characterized in that: Generate an inspection task list based on the risk level classification results and send it to the mobile terminal for execution, realizing dynamic regulation of security risks, including: Based on the risk level classification results, an inspection task list is generated, and inspection resources are allocated using a task priority sorting algorithm, wherein the inspection task list includes a first-level inspection task list, a second-level inspection task list, and a third-level inspection task list; the inspection task list includes inspection routes, monitoring point location coordinates, and inspection frequency requirements; If the security risk level is level 4, a third inspection task list is generated; if the security risk level is level 3, a second inspection task list is generated; if the security risk level is level 2, a first inspection task list is generated; Pushing the inspection task list to the mobile terminal device of the inspector to carry out the inspection work, and collecting image data and environmental parameters of the inspection points through the mobile terminal device, wherein the mobile terminal device plans the optimal inspection route based on the built-in electronic map navigation module; Analyze the image data and the environmental parameters, transmit the analysis results to the security and fire protection integrated big data platform, and output inspection feedback data; The parameters of the risk assessment model are updated through the inspection feedback data, and the risk level classification results are dynamically corrected.

3. The artificial intelligence-based security and fire protection integrated big data platform control method according to claim 2, characterized in that: The feature mapping matrix is ​​input into the constructed security assessment model, the security risk index is calculated, and the risk level is divided according to the security risk index, including: Based on historical data, the radial basis kernel function is used to construct a nonlinear mapping space, and the kernel function parameters and penalty factors are selected through cross-validation method to build a safety assessment model; Calculating weight coefficients for the spatial features and temporal features in the feature mapping matrix respectively, and constructing a composite feature vector by weighted combination; Inputting the composite feature vector into the security assessment model, calculating the distance from the sample point to the classification hyperplane, and obtaining the security risk index; The grading threshold of the security risk index is set according to the statistical distribution of historical data, and the changing trend of the security risk index at multiple consecutive time points is calculated using a sliding time window to establish the risk warning level and set the warning strategy.

4. The artificial intelligence-based security and fire protection integrated big data platform control method according to claim 3, characterized in that: The specific formula of the safety assessment model is as follows: Among them, R(t) is the safety risk index at time t, α i is the support vector coefficient of the i-th sample, x i is the support vector sample in the historical data, K(x i , H(t)) is the improved radial basis kernel function, H(t) is the feature mapping matrix, b is the bias term, λ is the regularization coefficient, w j is the weight coefficient of the jth feature dimension, F j (t) is the temporal feature integral term of the jth feature dimension, is the time window smoothing function, n is the number of samples, and m is the extracted feature dimension.

5. The artificial intelligence-based security and fire protection integrated big data platform control method according to claim 4, characterized in that: Also includes, When the security risk index R(t) < the first threshold and in the continuous time window |R(t k )-R(t k-1 )|<preset threshold, it is judged as level one risk; If the first threshold ≤ security risk index R(t) < the second threshold and the risk index R(t) at the current time k k )>The risk index R(t k-1 ), it is judged as a level 2 risk and triggers a level 1 warning; If the second threshold ≤ safety risk index R(t) < third threshold and the safety risk index R(t) remains in this range within the continuous time window, it is judged as a level 3 risk, triggering a level 2 warning, adjusting the inspection route, adding professional inspection personnel, and formulating an emergency plan; When the security risk index R(t) ≥ the third threshold or in the continuous time window |R(t k )-R(t k-1 )|>When the preset threshold is reached, it is judged as a level 4 risk, triggering a level 3 warning, activating the emergency response mechanism, and implementing special control measures.

6. The artificial intelligence-based security and fire protection integrated big data platform control method according to claim 4, characterized in that: The method for generating the feature mapping matrix is: Establish a deep convolutional neural network model based on the deep learning framework, and divide the training data set into training set and validation set according to the expected proportion; At the same time, a small batch stochastic gradient descent method is used to optimize the deep convolutional neural network model, wherein the deep convolutional neural network model includes an input layer, a multi-layer convolutional layer, a pooling layer, a fully connected layer and an output layer; The spatial correlation features are extracted through multi-layer convolution operations, the self-attention mechanism is introduced to calculate the weight coefficients between different channels, and the channel attention weight and the spatial attention weight are multiplied to obtain the fused attention weight; The fused attention weight and the original feature map are weighted and summed, and the weighted features are mapped to dimension reduction through a fully connected layer to generate a feature mapping matrix, wherein the dimension of the feature mapping matrix is ​​n×m, where n represents the number of samples and m represents the extracted feature dimension.

7. The artificial intelligence-based security and fire protection integrated big data platform control method according to claim 6, characterized in that: The method for generating the training data set is: Establish a multi-source data collection channel to collect the operation data of the integrated safety and firefighting big data platform through IoT sensors, where the operation data includes monitoring data, firefighting facility distribution data and safety inspection data; The monitoring data is cleaned to remove abnormal values ​​and duplicate values, and the missing values ​​are supplemented by linear interpolation method; At the same time, coordinate unification and scale standardization processing are performed on the fire protection facility distribution data, and time series alignment and format unification processing are performed on the safety inspection data; The maximum and minimum normalization method is used to map all processed data to the [0,1] interval to generate a training data set.

8. An artificial intelligence-based security and fire protection integrated big data platform control system, based on the artificial intelligence-based security and fire protection integrated big data platform control method according to any one of claims 1 to 7, characterized in that: include, An acquisition module is used to acquire the operating data and historical data of the security and fire protection integrated big data platform, and to standardize the operating data to generate a training data set; A feature extraction module, based on the training data set, trains and builds a deep convolutional neural network model, extracts spatial correlation features through multi-layer convolution operations, and combines the attention mechanism to identify key state information and generate a feature mapping matrix; A safety assessment and risk calculation module, used to input the feature mapping matrix into the constructed safety assessment model, calculate the safety risk index, and divide the risk level according to the safety risk index; The inspection task generation and distribution module is used to generate an inspection task list based on the risk level classification results, and distribute it to the mobile terminal for execution, thereby realizing dynamic regulation of security risks.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the artificial intelligence-based security and fire protection integrated big data platform control method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the artificial intelligence-based security and fire protection integrated big data platform control method described in any one of claims 1 to 7 are implemented.

Images